Know the MOE Syllabus | Secondary Computing 2027: G1, G2, G3, Programming, AI and Cybersecurity

Which 2027 Computing syllabus is my child taking?

The official SEAB directories list three distinct Singapore-Cambridge Secondary Education Certificate Computing subjects: G1 Computing K127, G2 Computing K237 and G3 Computing K349. G1 and G2 are described as new SEC subjects for 2027. G3 uses the new K349 code corresponding to the earlier O-Level reference code 7155. The course level matters because the assessed software, theory topics and relative importance of practical work differ.

Check SEAB G1, G2 or G3 according to the student’s registered subject. A child in Secondary 1 may still be developing foundational digital skills rather than preparing for an imminent national examination. The school determines its present teaching sequence, approved tools and available subject options under Full Subject-Based Banding.

Computing includes hardware, networks, programming and working responsibly with digital information. But the courses are not one interchangeable sequence of harder questions. For example, G1 covers document processing, spreadsheets, media software and programming; G2 emphasises spreadsheets, media and programming; G3 requires work involving Python and JupyterLab as well as broader theoretical concepts.

Before enrolling in a generic coding class, parents should ask which task the child currently cannot do independently. Is it understanding a question, building a logical sequence, reading program output, using spreadsheet references or identifying a security risk? Finding the exact bottleneck gives practice a purpose.

Back to contents

The 2027 Computing papers and their weightings

One of the most important differences between the subject levels is the assessment structure. G1 K127 has Paper 1 as a 1 hour 15 minute e-examination worth 40%, and Paper 2 as a two-hour lab examination worth 60%. The lab tasks cover media work, document processing and spreadsheets, and programming. The official G1 PDF is the best reference for required practical skills.

All candidates should check the software, permissions and examination arrangements supplied by their school. Success in a theory quiz is not proof of practical readiness, just as a learner who makes an impressive game may still need help with networks or data representation. Good preparation covers both knowledge and independent execution.

Back to contents

What computational thinking means in ordinary language

Computational thinking is the process of defining a problem so it can be solved through clear, testable steps. It includes decomposing a large problem, abstracting the relevant details, recognising patterns and building an algorithm. These habits apply even before the first line of code is written.

Imagine a school club organising a reading event. The organisers must collect names, check available places, assign time slots and send confirmations. A student could break the task into separate actions: accept details, validate them, compare remaining capacity, record decisions and generate appropriate messages. That decomposition makes the problem less overwhelming.

Abstraction asks what information is necessary. A participant’s name and selected time may matter. Their favourite movie may not. Removing irrelevant details reduces unnecessary work and helps protect privacy. Pattern recognition can reveal that every registration needs the same basic checks, while an algorithm specifies the order in which those checks occur.

Invite the learner to describe the procedure to a classmate who has never seen it. If the classmate cannot follow the instructions, the algorithm is not yet clear enough. This exercise can develop precision, teamwork and communication alongside computing. Computers operate reliably only when the instructions and assumptions are adequately specified.

Back to contents

Algorithms: a correct answer needs a reliable process

An algorithm is not simply any list of instructions. It must solve the intended problem under stated conditions, and the result should be testable. A program can display the correct output for one example while containing a flaw that appears with another input. Students should learn to distinguish a successful demonstration from a dependable solution.

Take a problem that asks for the highest score among five fictional students. A method could begin with the first score as the current highest and compare each remaining score, replacing the candidate whenever a larger score appears. The method works for different score values if the comparisons are correctly applied.

Now change the example. What if every score is negative, or two students have the same highest score? How should the program report a tie if the brief requires the names of all top scorers? The answers depend on what the problem asks, and the algorithm must be adjusted accordingly.

A useful study routine traces a small example by hand before running software. Write the input, current state and output at each step. That trace reveals incorrect assumptions and helps the student explain why the final answer follows from the process rather than from a lucky result.

Back to contents

G1 Computing: Scratch, graphics and practical digital documents

The 2027 G1 syllabus includes seven modules: computer fundamentals, networking, impact of computing, document processing, spreadsheets, media software and programming. Its programming learning outcomes include visual programming and creating simple games with sprites, events, coordinates and variables. These activities teach logical structure through direct interaction.

An original classroom game can ask a character to move to a target and increase the score only when contact occurs. The student needs to identify the trigger, the condition and the change in the score variable. If points increase without the character reaching the target, the learner should inspect which event is being detected.

Another practical task might involve producing a clear notice for a school event. A document needs suitable headings, readable layout and correct information. A spreadsheet could organise the event’s cost and attendance. A graphic or short video might communicate the main instruction. The lesson is not that every file must look elaborate; its purpose should remain understandable.

Parents should not dismiss visual programming as “just playing games”. Well-designed activities involve problem-solving and debugging. Equally, the child should not be forced to learn an advanced language before understanding sequences, conditions and loops. The G1 K127 syllabus provides an exact skills map.

Back to contents

G2 Computing: more independent work with media and data

G2 K237 consists of six modules: fundamentals, networking, impact, spreadsheets, media software and programming. Compared with G1, the separate document-processing module is not listed, while spreadsheet and computing tasks remain important. The actual syllabus sets detailed practical skills and the software environment for the examination.

Consider a group project analysing attendance over six weeks. The student enters figures in a spreadsheet, checks for typing errors, calculates totals and presents a chart. They should explain why the selected chart fits the data and whether the displayed trend actually supports the conclusion. A chart can be mathematically correct and still mislead through its scale or labels.

Media tasks add another dimension. The learner might create a short public-information clip using a storyboard, suitable graphics and clear text. The output should communicate efficiently to a particular audience. Flashy effects are not evidence of successful design if they obscure the information. The student needs to explain how the choices support the purpose.

In programming, the focus should remain on designing and testing behaviour, not copying a script that only works on the teacher’s example. Ask for one independent solution and one test case that could expose an error. That habit prepares students for practical assessment and for more sophisticated computing later.

Back to contents

G3 Computing: Python, JupyterLab and the underlying logic

G3 K349 includes computing fundamentals, algorithms and programming, spreadsheets, networking and the impact of computing. Its practical paper uses Python and JupyterLab. Students need to understand both the language constructs and the reasoning required to design, test and refine a solution.

A simple illustration is a program that calculates the total cost of items in a list. Before coding, the student should decide what each item contains, how a running total is updated and how the result should be presented. Then ask what happens if a price is accidentally entered as text or the list has no items. These details reveal why data types and validation matter.

Python syntax is useful, but a program that runs without an error is not necessarily correct. The student should compare its output with known examples, then try changed numbers and boundary conditions. If a bug appears, investigate the smallest section of code that could explain it rather than changing random lines repeatedly.

The official G3 K349 syllabus specifies the examinable skills, and the existing eduKate G3 Computing guide offers related explanations. Do not assume an advanced online course in artificial intelligence or web frameworks is automatically appropriate for SEC preparation. Master the registered fundamentals first.

Back to contents

Spreadsheets: formulas should update for the right reasons

Spreadsheets can store data, calculate results and create charts, but they require careful use of cell references. A student who copies a final number instead of using a formula may produce a correct-looking sheet that fails when inputs change. Computing involves understanding the relationship, not just using the interface.

Imagine a class selling tickets for a school event. One column records tickets sold and another the price per ticket. A third column should calculate revenue. When the number of tickets changes, the formula must update the total automatically. Ask the learner to explain which cells the formula uses and whether those cells represent compatible quantities.

Absolute and relative references are important when the same rate or condition applies across multiple rows. A formula that incorrectly shifts its reference may produce wrong outputs without displaying an error message. Students should test a copied formula on more than one row and explain which values are intended to remain fixed.

Charts need interpretation too. A pie chart may suit parts of a whole, while a line chart can show changes over time. A well-formed chart needs correct data, labels and scales. The learner should check whether the visual conclusion is supported rather than assume that software-generated graphics are automatically trustworthy.

Back to contents

Data representation: bits, files and meaningful information

Computers represent information in digital forms. Text, images, videos and numerical measurements are stored as data that software interprets. Students should distinguish the representation from the real-world thing it depicts. A digital photo is not the same as the event, and a file size is not the same as the quality or truth of its content.

The syllabus includes concepts relating to data units and computer hardware. A learner who confuses megabytes with gigabytes may misunderstand storage capacity. A student who treats image resolution as identical to physical size may create a poster that looks sharp on a phone but becomes blurred when printed. These are practical consequences of abstract definitions.

Raster images use pixels, while vector graphics describe shapes in ways that can often be resized more flexibly. Different file formats suit different purposes and may preserve or lose some information. Ask the learner why a logo, photograph and short video might use different file choices.

The objective is not to memorise every technical term without context. It is to develop a model of how data are stored, transformed and presented. That model helps learners make practical decisions and evaluate the reliability of digital outputs.

Back to contents

Hardware, software and the limits of specifications

A computer’s processor, memory and storage perform different functions. Students should not assume that a device with more storage is automatically faster at every task, or that a familiar brand guarantees performance. The actual workload, components and software matter.

For a school project involving simple document editing, the computer’s requirements may be modest. A task involving demanding graphics may place different demands on processing and memory. Comparing devices requires identifying the intended activity rather than simply listing the largest numbers in a product advertisement.

An operating system helps manage hardware resources and supports applications, while application software performs particular tasks such as spreadsheets or media editing. A student can demonstrate the difference by explaining why the operating system must remain available even when one particular editor is closed.

Ask students to think about maintenance and reliability as well. Safe updates, backups and appropriate software installations can prevent avoidable work loss. Good computing education develops an understanding of the whole system, not just isolated use of a favourite app.

Back to contents

Networks and the cloud: where does a shared file go?

Networks allow devices to exchange information and share resources. The Internet is a large interconnected system; a school’s local network is a more limited environment with its own rules. Cloud services rely on networks and remote computing resources rather than making information disappear into an undefined space.

Imagine a student saving a group assignment to a shared folder. If the file is stored only on the student’s laptop, other group members may not see the latest version. If it is stored through an approved online platform, access depends on permissions and connectivity. The students must know which version is current and how to recover from mistakes.

A useful exercise draws a diagram showing the user’s device, a network connection and a service holding the shared document. Ask what could fail at each stage. The file might be saved in the wrong folder, access permissions could block a classmate or the connection might be unavailable.

This lesson also introduces privacy. A public sharing link can give unintended people access, while an account with overly broad permissions may expose material that should remain private. Technical convenience and responsibility must be taught together.

Back to contents

Cybersecurity: protect accounts and information

The impact-of-computing topics include digital risks, such as scams, malicious software and improper disclosure of information. Students should understand common warning signs and defensive habits. The aim is to protect people, not to experiment with compromising someone else’s systems.

A message claiming that school access will be suspended unless the student immediately submits a password to an unfamiliar form is suspicious. The safer response is to use the school’s official sign-in route and ask a trusted administrator. A convincing logo or urgent wording is not proof that a message is genuine.

Other good habits include keeping approved software updated, using strong authentication, backing up important files and avoiding untrusted downloads. No single product guarantees complete protection. Learners should understand why multiple safeguards are used and why reporting a concern quickly is responsible.

Students should also think about other people’s privacy. A programming exercise rarely needs the real names, phone numbers or personal histories of classmates. Use invented or properly authorised data, apply appropriate sharing permissions and follow school rules for digital services. Computing education should make children more capable and more careful at the same time.

Back to contents

Artificial intelligence: critical use rather than automatic trust

AI systems can help generate suggestions, descriptions, code and images, but outputs can contain errors and may not disclose their limitations. The official syllabuses address the impact of computing and technology in ways suited to each subject level. Parents should not assume that every AI product or recent feature is an examined topic simply because it appears in the news.

A good classroom exercise presents a short AI-generated explanation and asks students to verify its claims against an approved source. Are the figures accurate? Does a cited link exist? Has the system invented an examination rule? The student records what was confirmed and what remains uncertain. This trains evidence-based judgement.

For coding, an AI assistant might suggest an algorithm, but the learner needs to understand and test it independently. A program that appears elegant may produce wrong results for unusual inputs or rely on a tool unavailable in the examination environment. Students should not surrender their ability to reason simply because output is generated quickly.

There are also privacy, fairness and authorship questions. Do not upload private information to unapproved AI systems or submit generated work as a record of independent practical assessment. The useful long-term outcome is a student who can use computing tools while remaining responsible for their own understanding.

Back to contents

Digital media: a useful product should have a clear purpose

G1 and G2 Computing include digital media skills. A student may work with vector graphics, raster images, presentations or short videos depending on the exact level. Success requires both using the software and making choices that help a viewer understand the intended message.

Suppose a class creates an information clip about a school recycling point. The storyboard should establish what the audience needs to know, show the relevant location in an appropriate way and make any essential instruction readable. A rapid series of transitions might look modern but fail if viewers cannot understand where to go or what to do.

Choose visual elements intentionally. A simple icon may be easier to recognise than a crowded photograph. Font size, contrast and spacing influence whether a message can be read on different devices. Each decision can be tested by showing the output to someone unfamiliar with the project and asking what they understood.

Students should use their own or appropriately licensed material. Obtaining a public image does not automatically mean it may be republished freely. Check school assessment rules and avoid uploading other people’s identifiable media without permission. Good digital creation respects both the audience and the creators of source material.

Back to contents

Debugging: find the first place where the logic breaks

Students who are anxious about programming errors may change many lines at once, hoping the program will suddenly work. A better routine is to identify the expected result, reproduce the error with a small test case and isolate the first step where actual behaviour differs. Fixing one cause helps learners understand the system.

For instance, a program is supposed to count completed homework submissions but reports one extra. The problem might arise because a header row is counted as a submission or an empty entry is treated as valid. Instead of guessing, the learner can trace each iteration and note when the counter increases. This shows whether the condition matches the intended definition.

After a correction, test again with a typical input and a different boundary case. A successful result for one dataset does not prove that the algorithm handles every relevant situation. Spreadsheet mistakes require the same discipline: check cell references, data types and assumptions before trusting an automated total.

Teachers can make debugging less threatening by treating errors as information about a model. A student who explains why a bug occurred is developing computational confidence. The aim is accuracy and independence, not merely to produce a screenshot without an error message.

Back to contents

Practical examinations: software proficiency includes file handling

The 2027 Computing lab papers require candidates to use approved school-provided software and submit the required work electronically. Practical confidence includes creating files, using relevant menus and shortcuts, saving in the required format and checking that the correct output was produced. A sophisticated answer may not be assessable if the student submits the wrong file.

A routine practice task should include the entire workflow. Read the instructions, plan the steps, create the artefact, verify the result and save the appropriate file. Leave a final minute to check names and locations when practising under suitable conditions. The student should follow the examination centre’s specific rules rather than an invented universal convention.

G1 and G2 practical assessments involve combinations of media, spreadsheet, document or programming tasks as listed in their syllabuses. G3 requires spreadsheet and Python/JupyterLab work. The school must confirm which approved versions and facilities candidates will use. Different software interfaces may produce similar outcomes but make practice less representative.

A parent can ask the learner to demonstrate a small task independently from a blank file. If the child knows the theory but relies on step-by-step teacher prompts for every action, practical fluency still needs development. Repeated, meaningful tasks are generally more valuable than watching many tutorials without creating anything.

Back to contents

Computing, Mathematics and Design: one problem, several skills

A practical computing problem may require mathematical accuracy, clear writing and sensible design. A programme that calculates an average needs correct arithmetic and a valid definition of the dataset. A user interface needs understandable labels. A digital tool designed for a group needs suitable assumptions about how people will use it.

Consider a simple budget tool for a fictional school club. Mathematics supplies quantities and percentages. Computing provides the data structure and formulas. Design thinking asks what information the user needs most clearly. The result is useful only when all three parts work together.

A student who gets every programming instruction right but misinterprets what the numbers mean can still produce a misleading tool. Another who chooses an elegant layout but does not validate inputs may give users unreliable results. The best projects encourage learners to test not only code but the quality of the underlying idea.

The MOE Mathematics 2027 guide and Design and Technology 2027 guide provide connected concepts. Students do not need to master a professional software-development pipeline at Secondary 1. They need to see that careful reasoning, evidence and communication belong together.

Back to contents

How parents can tell whether Computing tuition or practice is working

The number of hours spent in front of a computer is not a reliable measure of learning. A child may follow a long tutorial and produce a polished result while understanding very little of what they did. Another may spend half an hour solving one difficult logic error and make a much more significant advance.

Ask the learner to explain the task before starting, describe one important decision during the work, and demonstrate the result independently. If a teacher supplies a hint, note whether the child can solve a similar new problem without that hint. The ability to transfer a skill is more useful than immediate repetition.

Parents should also distinguish a software-interface problem from a conceptual one. A student might understand data analysis but not know where a spreadsheet feature is located; that requires tool practice. Another might use the feature effortlessly but reach an unjustified conclusion; that requires reasoning and interpretation. An effective lesson targets the correct cause.

The learner’s wellbeing matters. Building digital skills should not mean staying online until midnight or losing normal physical activity and friendships. A sustainable routine with clear targets, rest and sensible device use gives Computing room to grow alongside the wider curriculum.

Back to contents

A four-week SEC Computing revision plan

Week 1 — Confirm the actual subject: G1 K127, G2 K237 or G3 K349. Open the 2027 syllabus, check school software and collect one conceptual task and one practical output. Let the learner work without help first. Identify a narrow priority such as understanding loops, using cell references, media layout or cybersecurity judgement.

Week 2 — Repair the target. For logic, trace a small algorithm before coding. For spreadsheets, create a formula and vary the input. For media, build a short storyboard and test whether an unfamiliar viewer understands the message. For networks, draw a simple connection diagram and explain who can access the data.

Week 3 — Transfer to an unfamiliar scenario. Change the numbers, files or task instructions. Ask the student to predict an output and design a test that could reveal an error. Revisit an earlier problem after several days. If the learner requires repeated prompts, return to the underlying concept rather than adding harder work.

Week 4 — Practise a short theory and lab task aligned to the registered exam level, with a realistic file-saving check. Review mistakes by type and set a sensible next priority. Celebrate growing independence and accuracy, not simply the number of new programming tools installed.

Back to contents

Frequently asked questions

What are the 2027 SEC Computing codes?

G1 K127, G2 K237 and G3 K349. The registered subject level determines the official syllabus.

Are G1 and G2 Computing new SEC subjects?

Yes. SEAB identifies them as new in the 2027 school-candidate syllabus directories.

How much is the lab paper worth at G1?

The K127 lab paper contributes 60%, with the theory e-examination contributing 40%.

Is G2 assessment split equally?

Yes. G2 K237 uses a 50% e-examination and 50% practical laboratory paper.

Does G3 Computing use Python?

Yes. K349 practical assessment uses Python and JupyterLab alongside spreadsheet software, under official examination conditions.

Should G1 students learn advanced AI coding first?

No. Start with the appropriate syllabus and secure the prescribed digital, computational and programming foundations.

Can a working program still be wrong?

Yes. It may contain logical errors, unsuitable assumptions or results that fail when inputs change.

What is the best way to prepare for a practical test?

Practise the whole workflow using the correct school-approved tools: read, plan, create, test, save and verify.

Is cybersecurity only for computer specialists?

No. Protecting accounts, evaluating suspicious links and handling personal data responsibly matter to all learners.

Back to contents

Official Computing syllabuses and connected eduKate guides

• SEAB 2027 G1 Computing K127
• SEAB 2027 G2 Computing K237
• SEAB 2027 G3 Computing K349
• Official G1 K127 syllabus PDF
• Official G2 K237 syllabus PDF
• Official G3 K349 syllabus PDF
• SEAB SEC examination overview
• MOE secondary Full SBB curriculum directory
• eduKate G3 Computing syllabus guide
• Secondary Mathematics 2027: quantitative reasoning
• Design and Technology 2027: user needs and testing
• Economics and Business 2027: technology and financial literacy
• MOE and SEAB subject finder

Reviewed 6 November 2026. Exercises are original teaching examples rather than official SEC questions. Use the exact 2027 syllabus and school-issued candidate instructions for approved practical software and examination rules.

Back to contents